top of page

AI-Based Gamified Solutions for Children on the Autism Spectrum

Role: Research Assistant
Collaborators: Dr. S.N. Omkar (Supervisor), Dhruv Shinde (Collaborator)
Institution: Computer Intelligence Lab, Indian Institute of Science (IISc)
Duration: [June 2023 - December 2023]

Project Overview

Occupational therapists working with children on the autism spectrum often face engagement barriers, particularly with non verbal children. Therapy sessions targeting motor coordination, joint attention, and cognitive development require sustained focus, repetition, and individualized pacing.

​

This project explored how AI powered gamified systems could support therapists in improving gross motor skills and joint attention outcomes in non verbal autistic children.

​

Rather than building a finished product, the goal was to develop a research grounded intervention framework aligned with therapeutic practice.

Problem Context

Autism spectrum disorder affects approximately 1 in 36 children in the United States according to CDC estimates from 2023, based on surveillance data prior to that year.

​

In India, prevalence estimates vary, but community based studies suggest rates ranging between 1 to 1.5 percent of children (Raina et al., 2017; Indian Journal of Pediatrics).

Children with autism, particularly non verbal children, often experience:

  • Delayed gross motor development

  • Challenges in joint attention

  • Difficulty with Theory of Mind related tasks

  • Reduced engagement in structured therapy

​

Therapists report that maintaining motivation and repetition without overstimulation is one of the most difficult aspects of intervention delivery.

​

Core question:

How might AI supported gamified systems enhance therapist led interventions without replacing clinical expertise?

Research Objectives

  1. Understand real world occupational therapy workflows

  2. Identify friction points in therapy sessions

  3. Determine where gamification supports versus distracts

  4. Explore feasibility of AI driven adaptive systems

Key Findings

1. Engagement Drop Off Is Predictable

Children often disengaged during repetitive motor drills without visual or sensory reinforcement.

Opportunity: Introduce structured feedback loops that reward sustained movement patterns.

​

2. Overstimulation Is a Real Risk

Therapists cautioned that excessive visual or auditory feedback could dysregulate children.

Design implication: Minimalist, predictable reward systems are safer than high stimulation game mechanics.

​

3. Therapist Control Must Be Preserved

AI systems should support therapists, not automate therapy.

Control toggles and manual override features are essential.

​

4. Gross Motor Skills Were Under Digitally Supported

Most existing tools focused on communication apps. Fewer addressed structured gross motor reinforcement.

This informed the strategic pivot from speech therapy to motor skill development.

Methodology

Field Observation

​

Visited ASHA, a school for special children, to observe therapy sessions in naturalistic settings.

 

Focus areas:

  • Therapist child interaction patterns

  • Engagement breakdown moments

  • Instruction repetition cycles

  • Environmental triggers

Literature Review

​

Reviewed research on:

  • Joint attention interventions

  • Theory of Mind development

  • Gross motor skill acquisition

  • Digital assistive technologies in autism

 

Key insight:

Joint attention and motor engagement are foundational precursors to higher cognitive and communication gains.

Curriculum Segmentation

​

Analyzed therapy curriculum components and categorized them into:

  • Fine motor skills

  • Gross motor skills

  • Speech related interventions

 

This helped identify where AI based gamification would be developmentally appropriate vs disruptive.

Stakeholder Interviews

​

Conducted structured discussions with:

  • Occupational therapists

  • Special educators

  • Clinical experts at NIMHANS

 

Initial focus on AI driven speech therapy shifted based on expert feedback emphasizing greater need in gross motor intervention for non verbal children.

This pivot was evidence driven.

Design Direction

Based on synthesis, the proposed system included:

  • AI driven adaptive difficulty scaling

  • Real time motion detection feedback

  • Visual reinforcement for successful motor execution

  • Therapist dashboard for session customization

  • Low stimulation interface design

Gamification was structured around repetition tracking rather than competition.

Role

Researcher and Design Strategist. Responsibilities included:

  • Field observation

  • Expert consultation

  • Literature synthesis

  • Opportunity mapping

  • Intervention framework design

Impact

This project produced:

  • A validated pivot from speech focused to motor focused intervention

  • A structured opportunity map for AI in occupational therapy

  • A therapist informed feature blueprint

  • A research foundation for future prototype testing

  • Rather than launching a product, the outcome was a clinically aligned design framework grounded in practitioner realities.

Ethical and Clinical Guardrails

  • System positioned as assistive tool, not therapy replacement

  • Therapist in control at all times

  • Avoid overstimulation triggers

  • No autonomous diagnostic claims

  • Data privacy safeguards

Reflection

The most important shift was realizing that innovation in assistive care must begin with humility.

Gamification is not inherently beneficial. In sensitive populations, overstimulation or automation can cause harm.

This project reinforced that UX research in clinical contexts requires observation depth, stakeholder trust, and disciplined restraint.

bottom of page